Author + information
- Received January 31, 2017
- Revision received August 21, 2017
- Accepted August 22, 2017
- Published online November 29, 2017.
- Keith A. Dufendach, MDa,b,
- Katherine Timothy, BSc,
- Michael J. Ackerman, MD, PhDd,e,f,
- Benjamin Blevins, MDg,
- Andreas Pflaumer, MDh,
- Susan Etheridge, MDc,
- James Perry, MDi,
- Nico A. Blom, MDj,k,
- Joel Temple, MDl,
- Devyani Chowdhury, MDm,
- Jonathan R. Skinner, MBChB, DCH, MRCPn,
- Christopher Johnsrude, MDo,
- Andras Bratincsak, MDp,
- J. Martijn Bos, MD, PhDd,e,f and
- Maully Shah, MBBSb,∗ ()
- aDepartment of Pediatrics, Division of Pediatric Cardiology, Perelman School of Medicine at the University of Pennsylvania, Philadelphia, Pennsylvania
- bDepartment of Pediatrics, Division of Pediatric Cardiology, Children’s Hospital of Philadelphia, Philadelphia, Pennsylvania
- cDepartment of Pediatrics, Division of Pediatric Cardiology, University of Utah, Salt Lake City, Utah
- dDepartment of Cardiovascular Diseases, Division of Heart Rhythm Services, Mayo Clinic, Rochester, Minnesota
- eDepartment of Pediatrics, Division of Pediatric Cardiology, Mayo Clinic, Rochester, Minnesota
- fDepartment of Molecular Pharmacology & Experimental Therapeutics, Windland Smith Rice Sudden Death Genomics Laboratory, Mayo Clinic, Rochester, Minnesota
- gDepartment of Pediatrics, Division of Pediatric Cardiology, Naval Medical Center San Diego, San Diego, California
- hDepartment of Pediatrics, Division of Pediatric Cardiology, Royal Children’s Hospital, MCRI, and University of Melbourne, Melbourne, Australia
- iDepartment of Pediatrics, Division of Pediatric Cardiology, Rady Children’s Hospital/UC San Diego, San Diego, California
- jDepartment of Medicine, Division of Pediatric Cardiology, Leiden University Medical Center, Leiden, the Netherlands
- kDepartment of Medicine, Division of Pediatric Cardiology, Academic Medical Center, Amsterdam, the Netherlands
- lNemours/Alfred I. duPont Hospital for Children, Wilmington, Delaware
- mCardiology Care for Children, Lancaster, Pennsylvania
- nDepartment of Pediatrics, Division of Pediatric Cardiology, Starship Children's Hospital, Auckland, New Zealand
- oUniversity of Louisville, Louisville, Kentucky
- pPediatric and Adult Congenital Cardiology, Kapi'olani Medical Specialists, Honolulu, Hawaii
- ↵∗Address for correspondence:
Dr. Maully Shah, Perelman School of Medicine, University of Pennsylvania, The Children's Hospital of Philadelphia, 3401 Civic Center Boulevard, Philadelphia, Pennsylvania 19104.
Objectives The objective of this study was to evaluate contemporary clinical outcomes and identify triggers for arrhythmias or sudden death in an international cohort of Timothy Syndrome (TS) patients including those with novel TS-associated CACNA1C mutations.
Background TS is an extremely rare genetic disorder of the L-type cardiac channel Cav1.2 encoded by CACNA1C. The syndrome is characterized by multisystem abnormalities consisting of QT prolongation, congenital heart defects, syndactyly, facial dysmorphism, and neurological symptoms.
Methods Patients diagnosed with TS between January 1, 1994, and April 1, 2016, from 12 international tertiary care pediatric centers were included in this retrospective study. Data were gathered via survey from the patients’ electrophysiologists.
Results Seventeen patients diagnosed with TS were identified. Length of follow-up was 4.9 years (range: 3.0 to 19.0 years). Mean QTc was 640 ms (range: 500 to 976 ms). All patients were treated with beta-blockers; 13 patients (76%) were also treated with an implantable defibrillator. Eleven patients experienced an episode of aborted cardiac arrest, 6 associated with general anesthesia and 2 with hypoglycemia. Four patients died suddenly due to ventricular fibrillation, 2 of whom had associated hypoglycemia.
Conclusions This study shows that mortality in TS patients is due to multifactorial mechanisms, which include ventricular arrhythmias, pulseless electrical activity, and hypoglycemia. A simple nomenclature for ongoing studies of TS and related syndromes is described. A worldwide prospective registry is needed for continued exploration of this syndrome.
Timothy syndrome (TS) is an extremely rare genetic disorder characterized by myriad multisystem abnormalities, consisting of a cardiac phenotype that universally includes QT prolongation and potentially congenital heart disease (CHD) and/or cardiac hypertrophy, syndactyly, facial dysmorphism, and a neurological phenotype that can include autism, seizures, and intellectual disability. The original genetic discovery attributed the predominant cause to a recurrent de novo heterozygous missense mutation, p.Gly406Arg, in the alternatively spliced exon 8A of CACNA1C (1). Subsequently, 2 additional mutations were identified in exon 8 (p.Gly402Ser and p.Gly406Arg) associated with a slightly different phenotype than the original exon 8A-localizing missense mutation (2–4). With the advent of whole exome sequencing, other novel CACNA1C mutations have emerged, leading to an elucidation of phenotypic variants (5–11). Extreme QT prolongation is typical in TS and predisposes to ventricular fibrillation (VF), cardiac arrest (ACA), and sudden cardiac death (SCD). According to previous reports, most patients die before the third year of life despite medical treatment with beta-blockers (1). However, clinical outcome data reflecting current management strategies of this high-risk cohort have not been reported in the past decade.
We evaluated clinical outcomes and triggers for arrhythmia or sudden death in an international cohort of patients with TS, including those with novel TS-associated CACNA1C mutations. This case series provides a contemporary report of clinical course and outcomes in TS patients, along with updated management strategies and prognostic findings.
Patients diagnosed with TS from January 1, 1994, to April 1, 2016, from 12 international tertiary care pediatric centers were included in this retrospective study. De-identified data of TS patients from participating institutions were transferred electronically to the coordinating center and included baseline demographics, clinical characteristics, electrocardiogram (ECG) findings, genetic test results, treatment strategies, and outcomes. Additionally, hospitalization records including types of procedures and anesthetics were evaluated, along with the response and complications related to therapy.
Specific clinical characteristics evaluated included syndactyly, facial anomalies, abnormal dentition, baldness at birth, neurodevelopmental delay, CHD, suspected immune disorders, and hypoglycemia. The diagnosis of noncardiac manifestations, such as neurodevelopmental delay, suspected immune disorders, and hypoglycemia, were made based on the assessment by the patient’s treating clinician (12,13).
Patients were included in this study if they met the clinical criteria for TS with or without a confirmed genetic diagnosis as follows (1): QTc ≥480 ms and 1 or more of the following clinical features: 1) unilateral or bilateral cutaneous syndactyly; 2) typical TS facial and dental anomalies; 3) neurological symptoms including autism, seizures, intellectual disability, and hypotonia; and 4) CHD.
ECG and rhythm abnormalities
The standard pediatric ECG was reviewed for measurements and rhythm analysis. QTc was calculated according to the Bazett formula on a standard pediatric ECG. Maximum QTc recorded was used for analysis in this study. The T waves were analyzed for the presence of T-wave alternans (TWA). Bradyarrhythmias were defined as 1) pauses in the rhythm >3 s and 2) atrioventricular (AV) block. Tachyarrhythmias were classified as 1) torsades de pointes (TdP); 2) ventricular tachycardia (VT); 3) VF; and 4) atrial tachycardia/atrial fibrillation. Rhythm abnormalities data were obtained from documentation in medical charts, implantable cardioverter-defibrillator (ICD) analysis, and automatic external defibrillator (AED) rhythm analysis.
Genetic testing approaches included: 1) targeted analysis for pathogenic variants of CACNA1C, followed by whole gene sequence analysis if no pathogenic variant is found; 2) sequence analysis of CACNA1C followed by gene-targeted deletion/duplication analysis if no pathogenic variant is found; 3) use of a multigene panel that includes CACNA1C and other genes of interest; and 4) whole exome sequencing. Mosaicism was defined if the mutation had only a limited presence in the derived DNA, suggesting that 2 genetically distinct populations of cells coexisted within an individual’s somatic cells and/or gametes.
Cardiac implantable electronic devices
For patients with ICDs and pacemakers, device interrogation reports were analyzed for the presence of ventricular arrhythmias. The cardiac implantable electronic devices (CIEDs) were analyzed for the presence of: 1) TdP; 2) VT; 3) VF; and 4) supraventricular arrhythmias. ICD therapies were analyzed to determine whether therapies were appropriate and defibrillation successfully terminated the documented ventricular arrhythmia. Implantation of ICDs was classified as for either primary or secondary prevention based on whether they were implanted before (primary prevention) or after (secondary prevention) an ACA or documentation of sustained VT.
Data are expressed as mean ± SD, and median values are given when appropriate.
Seventeen patients (88% male) met study inclusion criteria. Age at diagnosis of TS was 17 ± 22 months (range: birth to 76 months). The median follow-up period was 4.9 years (range: 3.0 to 19.0 years), with a median age of 6.4 years at last follow-up.
Among all TS patients, a prenatal diagnosis of bradycardia secondary to functional 2:1 AV block was made in 4 patients (24%). Simple cutaneous syndactyly involving soft tissue only was present in 14 patients (82%) (Figure 1), neurodevelopmental delay in 11 (65%), facial anomalies in 8 (47%), abnormal dentition in 6 (35%) (Figure 2), baldness in 5 (29%), hypoglycemia in 7 (39%), CHD in 5 (29%), and suspected immunological disorder in 5 (29%). At last follow-up, 14 patients (77%) were alive. Patient characteristics are given in Table 1.
ECG and rhythm characteristics
Mean QTc values on the resting ECG were 640 ± 108 ms (range: 500 to 976 ms). TWA was present in 8 (47%), functional 2:1 AV block in 8 (47%), and bradycardia due to sinus pauses in 1 (6%), and was present at any time during the course of cardiac monitoring.
Details of patient mutations are summarized in Table 2. All 17 patients underwent genetic testing and had mutations in the CACNA1C-encoded cardiac calcium channel, although 1 was for research purposes and the results were not available. A missense mutation, p.Gly406Arg in the alternatively spliced exon 8 (exon 8A), was present in 10 patients (59%) (1–3). Three previously unreported mutations are detailed in this report. One mutation (p.Ser405Arg) in CACNA1C exon 8 was present in 2 related patients. Additionally, a p.Gly402Arg mutation in exon 8 was identified in 2 unrelated patients. Another novel mutation (p.Cys1021Arg) in exon 19 was identified in 1 patient with a classic TS phenotype of prolonged QT, TWA, and cutaneous syndactyly. In 2 children (12%), the TS-associated mutation was inherited from a parent with confirmed germline mosaicism. Of note, patient #12 had an additional mutation in KCNQ1 that was inherited from an asymptomatic father but was not diagnosed with concomitant LQT1. Finally, 7 of the 11 patients with ACA/SCD had exon 8A/8-localizing CACNA1C mutations.
All patients were treated with beta-blockers once the diagnosis of TS was clinically established (Table 2). Propranolol was used in 9 patients (53%), atenolol in 4 (24%), and nadolol in 4 (24%). Eight patients (#1, #2, #3, #5, #6, #7, #12, and #16) were taking beta-blockers at the time of the cardiac event (ACA or SCD). Beta-blocker therapy for Patient #2 was omitted on the day of his surgical procedure, and he had an ACA while under GA. Four patients (#1, #9, #10, and #13) presented with ACA and were diagnosed with TS after the event and therefore were not taking beta-blockers at the time of their cardiac event. These 5 patients were started on beta-blockers after the diagnosis of TS was established, and as noted earlier Patient #1 had another cardiac arrest while taking beta-blockers. Three patients were also treated with mexiletine. One patient had recurrent arrhythmia during a trial off mexiletine, so the medication was restarted.
A total of 13 patients (76%) had an ICD. ICD insertion was performed for primary prevention in 8 (61%) and secondary prevention in 5 (38%). ICD implantations were via a sternotomy in 6, with the high-voltage coil placed in the posterior pericardium and an abdominal ICD generator. Two patients had an ICD system that included an abdominal generator and a subcutaneous high-voltage coil, 4 had epicardial defibrillation patches with an abdominal generator, and 1 had a transvenous endocardial ICD system. No ICD complications were reported. Four patients (#5, #7, #13, and #16) had appropriate ICD discharges (3 VF, 1 TdP), which resulted in successful defibrillation in 2 patients (#7 and #13) and refractory VF leading to death in the other 2. The appropriate ICD discharge rate was 0.06/yr. No inappropriate ICD discharges were reported in the data. There were no ICD-related infections.
Four patients (#3, #5, #10, and #15) had a permanent pacemaker inserted for functional 2:1 AV block, 2 of whom (#5 and #10) went on to have an ICD inserted (1 for primary prevention and 1 for secondary prevention of SCD). There were no pacemaker-related infections.
Left cervical sympathetic denervation
Three patients (#3, #7, and #14) underwent left cervical sympathetic denervation (LCSD). The latter 2 patients also had an ICD inserted at the time of LCSD.
Tachyarrhythmias, ACA, and SCD
A total of 11 patients (65%) had ACA or SCD. Rhythm documentation was available in 10 patients (Table 3). One patient (#1) also had documented atrial fibrillation. The potential proarrhythmic triggers for the documented ventricular tachyarrhythmias were general anesthesia (GA) in 6, hypoglycemia in 2, and exertion in 1. SCD occurred in 4 patients (24%) (#3, #5, #6, and #16; Table 3), 2 of whom died despite an ICD. Notably, of those who experienced ventricular arrhythmia during or after GA, sevoflurane was used in at least 3, and 1 patient in particular has been arrhythmia-free in multiple other periods of GA by avoiding sevoflurane.
Four patients experienced SCD in this study. Patient #3 died at 1 month of life while still in the hospital after an episode of TdP. Patient #5, who had an ICD, developed refractory VF associated with deep hypoglycemia immediately after LCSD, which resulted in severe neurological damage and eventual discontinuation of treatment. Despite beta-blocker therapy and LCSD, this patient died of VF. Another patient (#6) initially had an ACA secondary to pulseless electrical activity, which was preceded by TWA during a pacemaker generator change under GA. The indication for permanent pacing was prolonged sinus pauses without 2:1 AV block. He was resuscitated successfully after 25 min with resultant significant anoxic neurological injury but did not have ventricular arrhythmias during the initial ACA event. The family did not agree to an ICD for this patient, who was then discharged home with an AED. One year later, after having made a nearly complete neurological recovery, the patient developed refractory VF while walking to the bathroom in the middle of the night and could not be successfully defibrillated with the family’s AED. The fourth patient (#16) died after a hypoglycemic event despite an ICD. At the time of hypoglycemia, the patient had a seizure, but the ICD recording did not show a ventricular arrhythmia. However, in the emergency department, the patient developed intractable VF and could not be resuscitated.
Although the first patients with marked QT prolongation and syndactyly were described in the 1990s (14,15), it was only in 2004 when the molecular genetic mechanisms and terminology of the disorder were first established (1). The complex clinical constellation of TS can be attributed to the ubiquitous presence of Cav1.2 throughout the body and alterations in channel function. In addition, decreased calcium-dependent and voltage-dependent inactivation of the Cav1.2 channel caused by the TS-associated mutations, calcium-independent effects, such as transcriptional modification, may contribute to the noncardiac, phenotypic manifestations of the syndrome (16–18).
There are 3 main findings in this study. First, beta-blockers as standalone therapy may not be protective against SCD. Also, all beta-blockers may not have the same efficacy. Regarding the choice of beta-blockers, data from single-center studies and expert consensus state that nadolol is the preferred antiarrhythmic drugs for treatment of long QT syndrome (LQTS). In this study, of the 8 patients who experienced a cardiac arrest while taking beta-blockers, 1 was treated with atenolol, 3 with nadolol, and 4 with propranolol. Unfortunately, nadolol is not available worldwide, and this may limit its universal use as encountered in this international study. There are no randomized clinical trials addressing the comparative efficacy of different beta-blockers for treatment of LQTS, but experts concur that when nadolol is not been available or tolerated, treatment with propranolol is preferred to beta1-selective agents such as atenolol (19).
Most patients were treated with a combination of antiarrhythmic agents and device therapy. At an average follow-up of 5 years, 77% of patients in our study were alive. These improved outcomes, compared with the initial report by Splawski et al. (1), may be attributed to early diagnosis and more aggressive medical and device therapy. Beta-blockers remain the mainstay of treatment, but mexiletine is an additional option for patients who continue to experience cardiac events while on beta-blockade. The evolution of smaller ICDs and alternative surgical implantation techniques has made it feasible to implant primary and secondary prevention ICDs in infants and young children, which was not possible in previous decades (20). Three patients also underwent LCSD, a procedure that has regained popularity in the past decade for primary and secondary prevention of malignant ventricular arrhythmias in LQTS (21,22). This remains an option for patients with TS, especially those with refractory symptoms.
The second finding of this study was that 8 of 11 ACA/SCD events (72%) were associated with either hypoglycemia or general anesthetic agents. It is known that patients with LQTS on beta-blocker therapy can develop episodes of hypoglycemia. We speculate that TS patients may be more prone to develop hypoglycemia due to potential episodic dysfunction of Cav1.2 signaling in the pancreatic beta-cells (1,4). This may be further complicated by treatment with beta-blockers, sometimes administered in high doses, which can mask early symptoms of hypoglycemia. Therefore, it would be prudent to avoid prolonged fasting in TS patients, and other strategies to prevent hypoglycemia should be investigated. GA, specifically with sevoflurane, was noted to be associated with ACA/SCD in several patients. Sevoflurane is known to prolong the QT interval even in normal children and may have further deleterious effects on patients with severe QT prolongation as seen in TS (23–25). Other contributory factors related to GA may include changes in sympathetic tone due to hypoxia, hypercarbia, local or light anesthesia, tracheal intubation, or emergence phenomena. Additionally, electrolyte imbalances and hypoglycemia may set in during the pre-procedural fasting period.
The third finding of this study was the prevalence of expanded TS-associated mutations within the CACNA1C gene and variable phenotypic features. Our study shows that the previously reported missense mutation p.Gly406Arg described in classic TS was present in only 10 patients (59%) (1–3). Three novel mutations were discovered in this case series (p.Ser405Arg, p.Gly402Arg, and p.Cys1021Arg). The clinical importance of this finding is that expanded genetic testing to the entire CACNA1C-encoded L-type calcium channel may be needed in patients with a TS-like phenotype if initial targeted testing of the hotspot exon8/8A is negative. Whether the variable TS phenotype is due to the type of mutation, its unique localization, alteration in signaling cascades, or other genetic/epigenetic factors is unresolved. Additionally, parental germline mosaicism was noted in 2 of our patients, which suggests some benefit for careful genotyping of the parent’s DNA even for parents without signs of TS (11,26,27).
Finally, it should be noted that our case series consists of a heterogeneous cohort of TS patients. None of the patients in our case series were presented in the original reports by Splawski et al. (1,2). In contrast, the original report by Splawski et al (1) published in 2004 described a cohort of 17 TS patients with molecular homogeneity, and all patients had syndactyly in addition to multiple organ system involvement. This is often referred to as TS type 1 or classic TS in the literature. In 2005, Splawski et al. (2) described 2 additional cases with extreme QT prolongation and arrhythmias but with the absence of syndactyly. Both cases had de novo missense mutations affecting exon 8 (p.Gly406Arg or p.Gly402Ser) but not exon 8A, leading to the coinage of the term TS type 2. In the ensuing 12 years, additional mutations in the CACNA1C gene on exons outside the hotspot region of exon 8A/8 associated with variable phenotypic TS features have been reported, making the task of further TS classification more complex. It may be prudent to formulate the TS nomenclature on the basis of the susceptibility genes rather than individual mutations or phenotypic characteristics. Therefore, we propose that all TS phenotypes resulting from mutations in CACNA1C be classified as TS1. Because every TS patient may not be positive for the CACNA1C mutation, when the next TS susceptibility gene is discovered, it can be classified as TS type 2. It is also important to recognize that CACNA1C mutations may result in a cardiac phenotype of QT prolongation, hypertrophic cardiomyopathy, CHD, and premature SCD without involvement of other organ systems. This should be annotated as cardiac-only TS (8). Another clarification needed is the terminology used for CACNA1C gene mutations resulting in isolated QT prolongation and arrhythmias without manifestation of other cardiac or extracardiac TS findings. Such cases should be simply referred to as LQTS type 8.
There are limitations to our data. Most notably, the small study cohort limits our ability to draw definitive conclusions regarding the different antiarrhythmic therapies prescribed. Given the retrospective nature of the study, clinical information on every aspect of multiorgan involvement in the cohort was not available. It is also possible that inclusion of novel CACNA1C mutations may have contributed to the positive outcomes in this cohort, although improvements in mortality were noted even among patients with the “classic” p.Gly406Arg mutations. Also, this study does not represent the complete list of TS patients diagnosed in the past decade. ICD complications were not encountered in this study, but this may be attributed to data collection bias and the small cohort of patients. In addition, although noncompliance with treatment with medications at the time of arrhythmic events was not reported by the treating physicians, the uniform compliance with treatment cannot be ascertained because of the intrinsic shortcomings of a retrospective study.
Our study shows that mortality in patients with TS is due to multifactorial mechanisms. Potential associations with sudden death events include hypoglycemia and specific general anesthetic agents. ICD therapy for primary and secondary prevention of SCD is frequently used for the management of TS with improved short-term survival. Due to the rarity of the disease, a worldwide prospective registry is needed for continued exploration into its clinical presentations, genetic mutations, malignant triggers, treatments, and outcomes.
COMPETENCY IN MEDICAL KNOWLEDGE: TS is an extremely rare genetic disorder of the L-type cardiac channel Cav1.2 encoded by CACNA1C. The syndrome is characterized by multisystem abnormalities consisting of QT prolongation, CHD, syndactyly, facial dysmorphism, and developmental delay. Early presentation with heart block due to QT prolongation is frequently seen, and a high attrition rate due to SCD has been reported in previous case series.
TRANSLATIONAL OUTLOOK: A worldwide prospective registry is needed for genetic and clinical exploration of this rare syndrome with multisystem abnormalities. Future studies are needed to identify potential triggers for SCD and optimal treatment strategies to prevent SCD.
This study was approved by the Pediatric and Congenital Electrophysiology Society (PACES). The views expressed in this article are those of the authors and do not reflect the official policy or position of the Department of the Navy, Department of Defense, or the United States Government. Dr. Ackerman has received consultant fees from Boston Scientific, Gilead Sciences, Medtronic, and St. Jude Medical; and royalties from Transgenomic. Dr. Shah has received a research grant unrelated to the study presented in this manuscript from Medtronic, Inc. Dr. Temple has served as a consultant for St. Jude Medical. All other authors have reported that they have no relationships relevant to this paper to disclose.
All authors attest they are in compliance with human studies committees and animal welfare regulations of the authors’ institutions and Food and Drug Administration guidelines, including patient consent where appropriate. For more information, visit the JACC: Clinical Electrophysiology author instructions page.
- Abbreviations and Acronyms
- aborted cardiac arrest
- automatic external defibrillator
- congenital heart disease
- cardiac implantable electronic device
- general anesthesia
- implantable cardioverter-defibrillator
- left cervical sympathetic denervation
- long QT syndrome
- sudden cardiac death
- torsades de pointes
- Timothy syndrome
- T-wave alternans
- ventricular fibrillation
- ventricular tachycardia
- Received January 31, 2017.
- Revision received August 21, 2017.
- Accepted August 22, 2017.
- 2017 American College of Cardiology Foundation
- Splawski I.,
- Timothy K.W.,
- Decher N.,
- et al.
- Diep V.,
- Seaver L.H.
- Betzenhauser M.J.,
- Pitt G.S.,
- Antzelevitch C.
- Boczek N.J.,
- Ye D.,
- Jun F.,
- et al.
- Boczek N.J.,
- Best J.M.,
- Tester D.J.,
- et al.
- Marks M.L.,
- Whisler S.L.,
- Clericuzio C.,
- et al.
- Phillip L.R.,
- Rodrriguez F.H. III.
- Hofmann F.,
- Flockerzi V.,
- Kahl S.,
- et al.
- Ackerman M.J.,
- Priori S.G.,
- Dubin A.M.,
- et al.
- Schwartz P.J.,
- Priori S.G.,
- Cerrone M.,
- et al.
- Bos J.M.,
- Bos K.M.,
- Johnson J.N.,
- et al.
- Schneider A.E.,
- Bos J.M.,
- Ackerman M.J.
- Dufendach K.A.,
- Giudicessi J.R.,
- Boczek N.J.,
- et al.